Infoscience

Thesis

Traduction des opérations de l'analyse historique dans le langage conceptuel des systèmes d'information géographique pour une exploration des processus morphologiques de la ville et du territoire

Present work intends to conceive an interface between Urban Morphology and Geographic Information Science (GISc). This interface is considered as a space of exchange for knowledge exploration between several disciplines. The links pointed out from this theoretical work help us define the framework of formalization of morphological processes. We have focused our efforts over the methodology of translation of the notion of morphological process. The main steps of translation of the historical analysis operations into the conceptual language of the GISc mainly include numerical handling of cartographic sources, i.e. maps, plans, cadastre, etc. The main problems encountered are linked to the heterogeneity of historical information such as: partial data sets, non-representative statistical ensembles and semantic relativity of concepts through time (polysemy, synonymy, etc.). We discuss these problems from the knowledge production process point of view. We focus on questions of knowledge emergence from quantitative analysis and, symmetrically, of concept formalization leading to data structures compatible with the idea of characterization. Our work is essentially based on the possibilities of the exploratory paradigm applied to geo-historical data. By doing so, we have been particularly interested in the contributions of the systematic approach to the enhancement of the interpretive potential of classical descriptive morphological analyses. Outcome of work are of different natures. We can mention among others: 1° the construction of a formal classification (ontology) of concepts subsuming transformation and evolution of the city and the territory and 2° the development of an exploratory approach of geo-historical data. We have also discussed the contribution of some social sciences' paradigms (Callon and Latour's sociology of translation, Koselleck's conceptual history (Begriffsgeschichte), Peirce's abduction, Ginzburg's indiciary paradigm, etc.) to GISc. In particular, we are interested in the enhancement of the process of modeling by rethinking the tasks of conceptualization and formalization by taking into account the historical context of production of information during the process of knowledge reduction. We have pointed out the importance of periods of conceptual stability to allow comparative measurements and of events to explain ruptures not only at the data level, but also at the level of the entire conceptualization. These outputs should be considered as a new step towards recognition of the exploratory approach as a relevant tool to face interdisciplinary and complex problems related to urban processes in general. In particular, this work helps us stabilizing the theoretical framework contributing to the definition of the next generation numerical archives and interoperability of systems. Indeed, these technical questions need modelers to take into account different kinds of evolution when dealing with either concepts or instances and to produce a common platform to manage both. Finally, we show some clues to continue this work by introducing collaborative approaches and exploratory multidimensional interfaces. The figure below illustrates the initial framework of this research and the extension resulting from our work.

Related material